On Mon, 10 Apr 2000, Holger Kruse wrote:
> The MiamiDx 1.0 release (complete with demo mode, docs and installation)
> is available now. Links are (intentionally) not on my web site yet,
Great! Perhaps this will stop some of the whining. :-)
BTW, perhaps the window title should change "Easy" to "Powerful" or
something in the Dx version. :-)
> - Bug fix: "MiamiNetStat -sL interface" would show bogus buffer overrun
> counts for interface that are offline.
Yeah, I'd noticed that one. :-) BTW, why does MiamiNetStat -I ignore the
numeric part of the interface names, and list ppp* or eth* or whatever?
> - Bug fix: Falling back to a default gateway with a lower priority
> did not work if the corresponding interface was a point-to-point
> interface.
Yes, that seems to work properly now. Unfortunately, the two "routing"
bugs related to determining the source IP are still present:
1) A loopback connection to the real local IP of a P2P interface uses the
IP of the default interface rather than the target interface, including
failing completely if the default interface is offline. This probably
doesn't affect many people, but I think I specified the bug and the fix
fairly precisely a while back.
2) NAT may choose the wrong source IP when multiple Internet interfaces
are online. I think it uses the IP of whichever Internet interface went
online earliest. This one is harder to figure out, but potentially
affects quite a few setups. AFAICT it's completely reproducible.
> - Added option to emulate the broken Windows-9x DHCP behavior (as
> expected by some Win-NT DHCP servers) of sending a user-defined
> pseudo-hostname in DHCP option 12 for ID purposes.
> - Removed all references to SAS/C time handling code and ENV:TZ.
> Modules that need to convert between local time and GMT now use
> locale.library.
Meaning that it no longer works properly under 2.04, the stated system
requirements notwithstanding?
> - MiamiTCPDump, miamipcap.library, miamibpf.library, MNI drivers:
> added option "-M" to connect MiamiTCPDump directly to an MNI
> driver. This allows all packet types (including non-IP/Arp types)
> to be dumped. MNI drivers were modified to support optional
> loopback of transmitted packets to listeners.
Hurray! Now there's a better tool to analyze all those buggy PPPoE
servers. :-)
BTW, is the lack of a 68000 version permanent?
I'll send some comments on the docs offlist.
Fred Wright
--
To unsubscribe send "unsubscribe miami-talk-ml" to
"[EMAIL PROTECTED]". For help on list commands send "help" to
"[EMAIL PROTECTED]".
